dbt Labs - Videos
Back to Channel
dbt MCP Server Fusion tool: Column-level lineage for AI agents
Column changes are one of the fastest ways to break things. The new dbt MCP Server Fusion tool helps your AI assistants and agents avoid that. It exposes dbt Fusion’s column-level lineage as a cal...
Using the dbt MCP Server Admin tool to automate dbt jobs with AI
Want AI assistants and agents to run your dbt Platform jobs and debug failed runs for you? Good news, you can. The dbt MCP Server Admin tool exposes jobs, runs, and artifacts as structured tools ...
Inside Snowflake’s AI roadmap (w/ Chris Child)
Snowflake VP of Product Management Chris Child joins Tristan Handy to unpack Snowflake's AI roadmap and what it means for data teams. They discuss the evolution from Snowpark to Cortex and Snowflak...
Compilation strategies in dbt
In this Fusion Fundamentals episode, Sara breaks down one of the biggest shifts introduced by the dbt Fusion engine: how compilation actually works. Fusion is a true SQL compiler that introduces a...
Build a CI agent with dbt structured context and MCP
In this demo, see how we built a CI Agent that uses dbt’s structured context layer to diagnose an issue, fix the model, and open a PR. End-to-end and fully automated. You’ll see how an agent resol...
dbt Community Awards 2025
Bolaji Oyejide, senior community manager at dbt Labs, hosts the 2025 dbt Community Awards at Coalesce 2025. Join the dbt Community: https://www.getdbt.com/community/join-the-community
Inside the dbt Fusion engine compiler
🧠 How does the dbt Fusion engine actually understand your SQL? In our latest Fusion Fundamentals video, we unpack what happens under the hood when Fusion compiles your dbt project. The key differ...
Speed, simplicity, cost savings: Experience the dbt Fusion engine
As analytics projects grow, so do the challenges: fragile pipelines, rising compute costs, and slow, error-prone development cycles. For Sonja Strempel, Analytics Engineer at DPG Media, the dbt Fu...
Building a multimodal lakehouse for AI (w/ Chang She)
In this episode of the Analytics Engineering Podcast, Tristan Handy sits down with Chang She — a co-creator of Pandas and now CEO of LanceDB — to explore the convergence of analytics and AI enginee...
What is SQL Comprehension?
Most SQL tools can render SQL, but they don’t really understand it. The dbt Fusion engine introduces SQL comprehension, a way for dbt to interpret SQL structure, semantics, and execution behavior ...
Exposing the dbt manifest through LLMs to make data discoverable (Coalesce 2025)
Using dbt metadata and LLMs, we built a system that helps users find and understand data assets beyond traditional search. Our solution creates a rich catalog focusing on business problems each mod...
Benn Stancil introduces a scorecard for measuring which robots work well with dbt (Coalesce 2025)
We built a new benchmark to test how well LLMs and AI products perform when trying to complete dozens of different analytics and data engineering tasks. This session will outline how the benchmark ...
Data that closes deals: Bridging GTM and analytics with dbt and AI (Coalesce 2025)
Go-to-market and data teams don’t always speak the same language. Conflicting priorities, unclear ownership, and limited context can slow progress, or worse, block it entirely. In this session, y...
Mesh around, find out: How to use dbt Mesh to enable business teams (Coalesce 2025)
This session will cover how Mercury Insurance implemented dbt Mesh, and how they work with their business-unit analytics teams to help them contribute to our mesh projects (from beginners to data p...
Building low-latency pipelines with dbt and Snowflake (Coalesce 2025)
Learn how Metropolis transformed a two-hour batch process into a low latency pipeline that identifies at-risk parking sessions while vehicles are still on-site, enabling immediate intervention and ...
How GitLab drove 95% documentation coverage in 4 days with Atlan, dbt, and AI (Coalesce 2025)
Among the most respected names in open-source software, GitLab’s 162 consecutive monthly releases have helped over 50 million users focus on shipping great code, faster. Key to delivering on this p...
Move from a legacy stack towards a cloud-native data and analytics platform (Coalesce 2025)
Learn how ING N.V. is redesigning its global data engineering and analytics operating model, merging centralized and decentralized approaches across federated business domains, while complying with...
dbt data migration: Best practices for large scale projects (Coalesce 2025)
In this session, attendees will learn best practices for migrating a large-scale dbt project from one platform to another, including running dual-execution environments, addressing syntactic differ...
Beyond the build: How PMG scaled dbt across clouds, teams, and impact (Coalesce 2025)
PMG scaled dbt from a scrappy Core setup in 2019 to a strategic Cloud investment—running across platforms, clouds, and teams. This shift standardized workflows, improved observability, and helped t...
Global X’s real-time asset management website with dbt Core, Etleap, & Snowflake (Coalesce 2025)
See how ETF provider GlobalX delivers fund data on its new website through Etleap ingestion and dbt Core modeling. One job pulls data from 14 market sources and auto-builds models for 100 funds whi...
FinAgent: Building autonomous AI agents for finance (Coalesce 2025)
How can AI agents make informed financial decisions—without compromising compliance, accuracy, or trust? In this session, we introduce FinAgent, a modular AI system that brings autonomy to high-sta...
From racetracks to casinos: How Churchill Downs is putting AI to work (Coalesce 2025)
At Churchill Downs, AI is enhancing everything from developer workflows to the betting window. In this session, you’ll learn how the team uses copilots to speed up BI development, RAG-based chatbot...
Testing: The SeQueL (Coalesce 2025)
We don't always have the luxury of bringing down prod when a critical test fails due to an expectation changing or some "bad" data getting through. We need a method to divert unexpected data into a...
How WHOOP unlocks smarter decisions with dbt and Snowflake (Coalesce 2025)
Join Snowflake and WHOOP to explore how product integrations with dbt are revolutionizing data transformation. They'll cover real-world business use cases we're helping our clients unlock. WHOOP wi...
How Zendesk built a cross-domain, multi-platform data strategy (Coalesce 2025)
Serving hundreds of engineers and analysts across dozens of domains takes more than good tooling. Learn how Zendesk scaled dbt across its distributed Snowflake architecture and multiple development...
Leveling up data engineering at Riot: How they transformed DevEx with dbt (Coalesce 2025)
Riot Games reduced its Databricks compute spend and accelerated development cycles by migrating from bespoke Databricks notebooks and Spark pipelines to a scalable, testable, and developer-friendly...
How trusted data powers growth at Ramp (Coalesce 2025)
Ramp is one of the fastest-growing software companies of all time, serving over 40,000 businesses with modern spend management. Their growth is fueled by the close partnership between product and d...
Governance as code: How dbt and AI unlock scalable data collaboration at ZScaler(Coalesce 2025)
As data teams embrace self-service, the need for strong, scalable governance becomes non-negotiable. Without clear guardrails, autonomy can quickly lead to chaos. In this session, you’ll learn how ...
From chaos to clarity: How Sweetgreen streamlined data with AI and dbt (Coalesce 2025)
In this session, you’ll learn how Sweetgreen is transforming its data ecosystem with dbt—enabling self-serve analytics and accelerating development. The team will walk through how they built a trus...
Getting buy-in for data infrastructure development (Coalesce 2025)
Data infrastructure development is important! But it can be difficult for business stakeholders to correctly prioritize it alongside everything else. In this talk, they'll go over some strategies f...
Building long-horizon agent workflows with OpenAI and dbt
In this session, you’ll see how OpenAI and dbt are partnering to make business-ready AI a reality. By combining OpenAI's GPT-5 based agents with the dbt Semantic Layer, you can build agents that re...
Cutting costs and time with smart dbt macros at Delivery Hero (Coalesce 2025)
Data teams often face trade-offs between model freshness and compute cost. Delivery Hero built dynamic backfill macros to enhance their incremental models, enabling their teams to refresh data effi...
Why self-serve analytics and AI fail and how metadata can save them (Coalesce 2025)
Self-serve analytics promises speed. But without clear guidance, it often leads to hidden obstacles like cluttered dashboards, runaway costs, and a loss of trust in the data. Add AI to the mix, and...
So you want to build a data mesh (Coalesce 2025)
Let's get meshy! Learn how to evolve your organization's dbt project(s) to more effectively manage distributed data at scale by adopting data mesh as a grounding philosophy. This is your roadmap of...
Goodbye manual testing & alert fatigue: Meet your AI data SRE (Coalesce 2025)
Eliminate 80% of the manual effort spent writing dbt tests, chasing noisy alerts, and fixing data issues. In this session, you'll see how data teams are using an AI Data SRE that detects, triages...
Freedom through structure: How WHOOP scales analyst autonomy with dbt (Coalesce 2025)
AI and dbt unlocks the potential for any data analyst to work like full-stack dbt developers. But without the right guardrails, that freedom can quickly turn into chaos and technical debt. At WHOOP...
How Wikimedia reduced reporting latency 10x using dbt and Iceberg (Coalesce 2025)
Learn how the Wikimedia Foundation implemented an on-prem, open source data lake to fund Wikipedia and the future of open knowledge. They'll discuss data architecture including challenges integrati...
How Virgin Media O2 built scalable self-service with dbt across two orgs (Coalesce 2025)
Merging two large organizations with different tools, teams, and data practices is never simple. At Virgin Media O2, they used dbt to help bring consistency to that complexity, building a hybrid da...
Cost-effective data operations (Coalesce 2025)
Delivering trusted data at scale doesn’t have to mean ballooning costs or endless rework. In this session, we’ll explore how state-aware orchestration, powered by Fusion, drives leaner, smarter pip...
Mamma mia! My data’s in the Iceberg (Coalesce 2025)
Iceberg is an open storage format for large analytical datasets that is now interoperable with most modern data platforms. But the setup is complicated, and caveats abound. Jeremy Cohen will tour ...
Unleash the power of dbt on Google Cloud: BigQuery, Iceberg, DataFrames and beyond (Coalesce 2025)
The data world has long been divided, with data engineers and data scientists working in silos. This fragmentation creates a long, difficult journey from raw data to machine learning models. We've ...
Building high-quality AI workflows with the dbt MCP Server (Coalesce 2025)
Modern data teams are tasked with bringing AI into the analytics stack in a way that is trustworthy, scalable, and deeply integrated. In this session, we’ll show how the dbt MCP server connects LLM...
What’s new in the dbt language across Core and Fusion (Coalesce 2025)
The dbt language is growing to support new workflows across both dbt Core and the dbt Fusion engine. In this session, we’ll walk through the latest updates to dbt—from sample mode to iceberg cata...
Towards a more perfect pipeline: CI/CD in the dbt Platform (Coalesce 2025)
In this session, the United Services Automobile Association will show how they integrated CI/CD dbt jobs to validate data and run tests on every merge request. Attendees will walk away with a bluep...
Analysts 2049: Redefining the role of the Analyst in the age of AI (Coalesce 2025)
In this session, you’ll learn how dbt’s Semantic Layer and MCP Server enable safe, AI-powered access to data through natural language interfaces. As business users start exploring data conversation...
7 ways Bumble reduced its tech debt and prepared for ML/AI (Coalesce 2025)
Setting up a dbt project the right way ensures maintainability, scalability, and efficiency. Learn how Bumble uses dbt to reduce tech debt and lay the foundation not only for analytics, but also fo...
How Mammoth Growth built a multi-agent dbt workflow (and crushed our backlog) (Coalesce 2025)
Manual modeling is out. Multi-agent systems are in. In this session, Dylan shows how his team at Mammoth Growth paired dbt with autonomous agents to handle complex modeling tasks. The result was ...
Feature feast: Building a data buffet for models (Coalesce 2025)
Fifth Third Bank transformed its MLOps using a feature store built with dbt Cloud. The result? Improved model governance, reduced risk and faster product innovation. In this session, learn how they...
From one-off wins to org-wide impact: Scaling AI at J.Crew (Coalesce 2025)
AI isn’t just a lab experiment at J.Crew. It’s already driving results. This session unpacks how a few early wins with AI are being scaled across the company. You'll learn how to spot issues earl...
Building bridges: How Roche GTM used dbt to earn stakeholder trust (Coalesce 2025)
Building stakeholder trust through data is crucial for organizations. At Roche, we used dbt to create a transparent, robust data transformation process that helped gain confidence and increase dash...